126
Biochimica ei BiopltysicaActa, 1107(1992) 126-130 © 1992ElsevierScience PublishersB.V.All rightsreserved0005-2736/92/$05.00
Proton potential-dependent polyamine transport by vacuolar membrane vesicles of Saccharomyces cerevisiae Yoshimi Kakinuma, Naoyuki Maqttda and Kazuei Igarashi Facultyof PharmaceuticalSciences, Chiba Unirersity, Chiba (Japan)
(Received21 January 1992)
Key words: Polyaminetransport; Proton potential; Vacuolarmembranevesicle;(S. cereridae) Vacuolar membrane vesicles of Saccharomyces cerevisiae accumulated spermine and spermidme in the presence of ATP, not in the presence of ADP. Spermine and spermidine transport at pH 7.4 showed saturation kinetics with K m values of 0.2 mM and 0.7 raM, respectively. Spermine uptake was competitively inhibited by spermidine and putrescine, but was not affected by seven amino acids, substrates of active transport systems of vacuolar membrane. Spermine transport was inhibited by the H +-ATPasespecific inhibtors bafilomycin A I and N,N'-dicyclohexylcarbodiimide, but not by vanadate. It was also sensitive to Cu 2+ or Zn 2+ ions, inhibitors of vacuolar H +-ATPase. Both 3,5-di.tert.butyl-4-hydroxTbenzilidenemalononitrile (SF6847) and nigericin blocked completely the spermine uptake, but valinomyein did not. [14C]Sp~.rmine accumulatcd in the vesicles was exchangeable with unlabeled spermine and spermidine. However, it was released by a protonoph~re only in the presence of a counterion such as Ca z+. These results indicate that a polyamine-specific transport system depending on a proton potential functions in the vacuolar membrane of this organism.
Introduction Polyamin~.s, represented by putrescine, spermidine and spermine, are found to be widely distributed from bacteria through to higher organisms [1]. The significance of these polyamines in cell proliferation is clear, since the depletion of intraceUular polyamines by the blockage of synthetic pathways with inhibitors or the mutation of polyamine biosynthetic enzymes halts the cell growth. It can then be restored by the addition of polyamines to the medium [2]. The cellular polyamine level is usually regulated by both its synthetic pathway via ornithine decarboxylase as the rate-limiting enzyme with a very short half-life and by its degradative pathway [2]. It has recently become accepted that movements of polyamines across the cell membrane are also important for regulation of its cellular content [3]. Uptake systems specific for polyamines are reported in bacteria and animal cells Correspondence: K. Igarash!, Faculty of Pharmaceutical Sciences, Chiba University,1-33 Yayoi-cho,Chiba 263, Japan. Abbreviations: ANS, l-anilinonaphthalene.8-sulfonicacid; CCCP, carbonyl cyanide m-chlorophenylhydrazone;DCCD. N,N'-dicyclohexylcarbodiimide;HPLC. high-performanceliquid ¢.hromalography; Mes, 2-(N-rnorpholino)ethanesulfonicacid; SF6847, 3,.~-di-tert-butyl4-hydroxybenzilidenemalononitrile.
[4-7]; we found three polyamine uptake systems, which depend on the proton potential, in Escherichia coil [8,9], and a membrane potential-dependent uptake system of polyamines in bovine lymphocytes [10]. As polyamines and their derivatives synthesized in vivo are released into the medium [11], their extrusion pathways must exist. As polyamines easily bind to cellula~ anionic constituents such as RNA, DNA and phospholipids [12], it is hard to define the target site(s) crucial for cellular physiology. However, we understand that binding of polyamine to the ribosomal machinery or m R N A plays an important role in protein synthesis in the cytoplasm [13,14]. The cytoplasmic polyamine level is probably influenced by its compartmentalization in various organelles of eukaryotes: a few reports have indicated the existence of the polyamine transport system in mitochondria [15,16]. Vacuoles compose the largest compartment in plants and fungi [17]. It has been proposed that vacuoles maintain cytoplasmic homeostasis of nutrients and ions and act as a metabolic compartment or lytic compartment in intracellular digestive processes [17]. In Neurospora crassa, it was reported that spermidine is localized in vacuoles [18]. However, the mechanism of polyamine accumulation into vacuoles has ,,ot been investigated.
127 We examined here polyamine transport by vacuolar membrane vesicles of Sa,echaromyces cerevisiae, and found a polyamine specific transport system depending on a proton potential. Materials and Methods
Strain and culture condition. All the experiments were conducted with Saccharomyces cerevisiae haploid strain X2180-1A from the Yeast Genetic Stock Center, Berkeley, CA, USA. Cells were grown in medium (YEPD) containing 1% Difco yeast extract, 2% polypeptone and 2% glucose at 30°C with subsequent shaking at 120 strokes per rain. Chemicals. [t4C]Putrcscin¢, spermidin¢ and spermine (4 G B q / m m o l ) were purchased from DuPontNew England Nuc~.ear. SF6847 was a gift from Dr. Y. Anraku, University of Tokyo, Japan. Other chemicals used were of analytical grade. Preparation of spheroplast lysate, intact vacuoles and right-side-out vacuolar membrane vesicles. Spheroplast lysate, intact vacuoles and right-side-out vacuolar membrane vesicles, which were essentially free of mitochondrial contamination (no succinat¢ dehydrogenas¢ activity as a marker enzyme for mitochondria), were prepared by the method of Ohsumi and Anraku [19]: in short, intact vacuoles were released into the lysat¢ by an 'osmotic shock' disruption of spheroplasts, and then separated from other organelles and cellular constituents by centrifugation. Only intact vacuoles are floated in 8% Ficoll buffer during centrifugation at 2 6 6 0 0 × g for 30 rain (see rgf. 19 for details). The vesicles were perfectly spherical and 0.2-1.0 /~m in diameter. An internal water space of vacuolar membrane vesicles of 5.2 p.I/mg membrane protein [20] was used for calculation of the concentration. Transport assay. The standard assay mixture (100 btl) consisted 6f 25 mM Tris-Mes (pH 7.4), 4 mM MgCI 2, 25 mM KCI, 0.5 mM ATP, 0.1 mM polyamine (74 MBq/mmol) and 30-50 p.g protein of vacuolar membrane vesicles. The reaction at 25°C was initiated by adding labeled substrate and :~rminated by diluting the reaction mixture with 5 mi of cold buffer consisting of 10 mM.Tris Mes (pH 6.9), $ mM MgCI 2 and 25 mM KCI. The vacuolar membrane vesicles were recovered on a membrane filter (Millipore, 0.45/tm) and washed with 5 ml of the above buffer. The radioactivity was determined with a liquid scintillation counter. Polyamine analysis. Polyamine contents in spheroplast lysates, intact vacuoles and vacuolar membrane vesicles were determined with HPLC as described elsewhere [21] after extraction with hot triehloroacetic acid, Other procedures. Protein was determined by the method of Lowry et al. [22] with bovine serum albumin as standard.
TABLE 1 Retentionof polyamim's it! cacuoles S. cerevisiaeX2180.1Agrown on YEPD mediumwere harvestedat 3' 107cells/ml, and then spheroplastlysateand intact vacuoleswere prepared as described in Materials and Methods. Intact vacuoles were then disrupted by osmotic shock to convert to membrane vesicles, and the vesiclesand their supernatant were recovered by centrifugation[191.These fractionswere extracted twice with 0.5 ml of 5% Iric~ ...'oacetic acid, and aliquots were then analyzed for polyaminesby HPLC as described previously[21]. Standard error was within the range of 5% in triplicate experiments. PUT, putrescine; SPD, spermidine; SP. sperrnine. Numbers in parentheses show total polyaminecontent (in nmoDin each f:'sction: n.d., not detected. Fraction
Pmtein nmol/mgpmtein(nmol) (mg) PUT SPD SP
Sphcroplast lysate
238
Intact vacuoles 1.7 After precipitation Precipitate(vesicles) 0.4 Supernatant 1.5
3.9
51.6
0.8(1.4) ~.6(103) n.d, 1.1(I.7)
0.5 1.9(3.2)
72.5(~) 0,9(0.4) 43.3(65) 1.9(2.9)
Results
Retention of polyamines in vacuoles Table I shows the polyaminc contents of spheroplast lysate, intact vacuoles and vacuolar membrane vesicles. Strain X2180-1A was grown on YEPD medium, harvested at 3" 107 cells/ml, and individual fractions were prepared as described in Materials and Methods. In the fraction of purified vacuoles, putrcscine, spermidine and spermine were all detected, although spermidine was preferentially presented. By osmotic disrup, lion, 100% of putrcsin¢, 65% of sperrnidin¢ and 90% of spermine in the vacuoles were released, indicating that all these polyamines were actually retained in vacuoles. Based on a-mannosidase activity as a specific marker enzyme of the vacuolar membrane [20], about 13% of total vacuoles in the spheroplast lysate are recovered by this purification method. This means that 1% of putrescine, 6% of spermidine and 21% of spcrmine of the lysat¢ are localized in vacuoles although the possibility of a leakage of polyamine from vacuoles by this osmotic shock method cannot be ruled out (Table 1): it was reported that about 30% of cellular spermidine is localized in vacuoles of Neurospora crassa
[181. ATP-dependent polyamine uptake by vacuolar mere. brahe ve,~icles Anraku and his colleagues have been extensively studying the energetics of yeast vacuoles, and have demonstrated ATP-driven transport systems of Ca 2+ and amino acids [19,23,24]. All of these systems are actually driven by a proton potential (about 180 mV,
128 TABLE II 10!!
.
.
.
.
.
.
.
.
.
Kineticparametersof A TP-rh,pendent pnlyammeuptake
.
Uptake assayswere performed as described in Materialsand Methods. Kinetic constants were estimated from the initial velocityof uptakes.The Ki valuesof competitiveinhibilkmweredeterminedby varying the substrate concentrafion in the presence of a fixed inhibitoreonceutration.Standard error was will|inthe rangeof I(1% ill
iF
triplicate experiments. SP, sperminc; SPD, spermidine; PUT, pu.
trescine. Substrate
Km (mM)
Putrescine Spermidinc Spermine
0
0
"
"
2
"
' 4
6
8
10
(rim)
Fig. I. Time courseof ATP-dependent polyamineuptakeby wteuolar membrane vesicles.Uptakes of spermine (el. spermidine ( • ) and putreseine (Ill) into the vesicleswere assayed under standard conditions in the presence of 11.5 mM ATP. Each point represents the average of triplicate experiments. Standard error was within the range of I()q~:.
intravaeuolar positive) generated by a vacuolar H +translocating ATPase, and are suggested to be an n H + / C a -'÷ antiporter or n H + / a m i n o acid antiporters [19,23,24]. We examined the effect of ATP on accumulations of [t4C]putrescine, spermidine and spermine into vacuolar membrane vesicles (Fig. I). The uptake activities of spermidine and spermine were greatly stimulated by ATP, but the activity of ATP-dependent putresine uptake was very low. In the absence of ATP, no significant uptake of polyamines was observed. At 5 min after the addition of ATP, established concentration gradients of spermidine and spermine across the vacuolar membrane were approximately 4 and 17, respectively, indicating ATP-dependent active transport of these polyamines by vacuolar membrane vesicles. When the experiments were performed at 0°C, no accumulation of these polyamines was observed even in the presence of ATP.
Specificio of the uptake system Table 11 shows the kinetic parameters of ATP-dependent polyamine uptake by vesicles. Considering the values of K m and Vma~ for the uptake activities of putreseine, spermidine and spermine (Fig. 1), spermine was the substrate with the highest affinity. Spermidine uptake was inhibited by putrescine (data not shown) and spermine, and spermine uptake was inhibited by spermidine and putrescine. Inhibitions by these
2.0 0.7 0.2
Vma~ (nmolmin t mg protein)- i 0.8 6 7
Ki (mM)
0.1 (SP) 0 8 (SPD), 1.1 (PUT)
polyamines were apparently all competitive; the K~ values of these polyamines are shown in Table ll. Eight transport systems of amino acid are found in vacuolar membrane [24]. Seven of these are eriven by a proton potential generated by H+-ATPase [20]. To determine whether polyamine uptake is mediated by a specific transport system distinct from these systems, the effect of these seven amino acids on spermine uptake by ve~eles was examined (Table liD. Sperm(he uptake was hardly inhibited by the addition of these amino acids, suggesting that an ATP-dependent transport system specific for polyamines exists on the vacuolar membrane.
Properties of polyamhw uptake Table IV shows the nucleotide specificity of spermine uptake. Spermine transport was not only stimulated by ATP but also by GTP, UTP and CTP in this order. No transport of spermine was observed by the addition of ADP. This preference corresponded well to the nueleotide specificity of the vacuolar membrane H+-ATPase [20]. The effect of various ATPase in-
TABLE III
I'~ffi'clof rut(pus L-ambzoachlson ATP.dept'ndentspernlbwuptake The initial rates of sperm(he uptake were determined under star,. dard condithms. The concentrationsof labeled spermine and unlabeled amino acid were 0.1 and 2 raM. respectively.Each point represents tbs mean of duplicate experiments. Aminoacid
Relative activily (%)
None Arginine Lysinc Histidinc Plmnylalanine Tyrosiue Glutamine lsoleucine
I(}0 102 83 138 99 84 l Ir~ 127
129 TABLE IV NucleotMe.~pecificityof spermine uptake Initial rates of spermine uptake were determined uuder standard conditions.The concentrationof nucleotide was 0.4 raM. Each point represents tile mean of duplicate experiments.
GTP UTP CTP ADP
I:'
.
.
.
.
.
•
.
.
.
.
°
o
I.
86 45 32 0
i,
hibitors and ionophores on ATP-dependent spermine uptake was also examined (Table V). Vesicles preincubated with 10 p.M bafilomycin A t [25] or 0.1 mM N,N'-dicyclohexylcarbodiimide, a specific inhibitor for the vacuolar H +-ATPase, showed no spermine uptake activity. By contrast, vanadate, an inhibitor of the ErE 2 type ATPase, slightly activated spermine uptake. Cupric and zinc ions, which strongly inhibit H+-ATPase [20], blocked the activity completely. Calcium ions also inhibited the uptake, but to a lesser degree than Cu 2+ and Zn 2+. When magnesium ions were removed from the reaction mixture, the uptake activity was completely, lost, indicating that they are essential for the uptake activity. These results suggest that the vacuolar [-I+-ATPase participates in the spermine uptake by membrane vesicles. Moreover, spermine uptake was severely inhibited b~ the protonophores SF6847 and CCCP, as well as by an electroneutral K + / H + an-
TABLE V
Effi'ct of carious reagentson ,4 TP.dependent Sl~'nnhwuptake Assayswere p~:rformedunder standard conditions and the activity without addition (control) w::s taken as 100%. Membrane vesicles were preincuhatedwith bafilomyeinA I or DCCDat 25°Cfor 10 rain before assay. Other rcageats were added just before assay. Each point representsthe mean of duplicate experiments.
Addition
1 0 • A.
Cunceati'ation (mM)
Relativeactivity (%)
0.01 O.2 0.1
9 0 152
MgCI~
0.I
I0{}
CaCI: CuCI2 ZnCI2
0.1 0.1 0.1
55 0
~
(ml,)
Fig. 2. Effluxof [t4Clspcrminefromvacuolarmembranevesicles.(A) Effect of ur.labeled polyamine.ATP-drivenspermineuptake (o) was carried out un..ter standard conditionswith 0.1 mM 114C]spermine. At the arrow, 1 mM spermine to) or 2 mM spermidine (Ill) was added to portionsof the reaction mixture.(B) Effect of 81:6847 and Ca:+, ATP.driven spermineuptake (0) was carried out as the same as in Fig. 2A. At the arrow.2/zM SF6847(o) or SF6847plus 2 mM CaCI2( • ) was added to portionsof the reaction mixture.
tiporter, nigericin, indicating that it depends on a proton potential. However, valinomycin did not inhibit spermine uptake activity (Table V). The same effect of these reagents on ATP-dependcnt spermidine uptake was obtained (data not shown). All these results suggest that active polyamine transport by membrane vesicles is driven by a proton potential generated by the vacuolar H %ATPase. To determine whether accumulated spermine in vesicles is in a free, osmotically active form, the effiux of spormine was examined (Fig. 2). The spermine accumulated in the vesicles was exchangeable with a 10-fold excess of nonradioactive spermine (Fig, 2A). Spermidine also prompted the exchange diffusion of spermine, but to a lesser extent. Howew:r, spermine was not in a completely free form, since SF6847 did not readily release it (Fig. 2B). When Ca 2+ ions were added as eonv.ter ion to the suspension, the effiux of spermine became accelerated (Fig, 2B).
ATPase inhibito~
BafilomycinA t DCCD Sodiumvanadate Divalent cations
Ionophores SF6847 CCCP
0
0.001
12
Nigericin
O,D2 0.O01
6 23
Valinomycin
0.01
98
Discussion
In this work, we used vacuolar membrane vesicles in which the intravacuolar polyamines were depleted (Table I), instead of the fragile vacuoles, and found a new active transport system of polyamines in vacuoles of S. cerevisiae. This system was ATP-dependent and polyamine-specific. Judging from the sensitivity of spermine uptake to various ATPase inhibitors and ionophores, this system depends on a proton potential generated by H+-ATPase [20]. We expect that the polyamine uptake system may be a substrate/nH +
130 antiporter as reported in the uptake systems of arginine and Ca 2+ [19,23]. However, we must consider that accumulated spermine in vesicles is probably not in a free form; spermine at a steady-state level was not released even when the proton potential was dissipated by the addition of the protonophore SF6847 (Fig. 2B). Accumulated sperreine probably binds with some anionic constituents in the vesicles. Diirr et al. [26] measured the amounts of amino aCids and polyphosphates in the vacuoles of yeast, and found a linear correlation of their contents, from which they suggested that polypbosphates act as ion exchange resins to conserve amino acids stably in the vacuolar space [26]. Also in the case of N. crassa, it is pointed out that polyphosphates sequester spermidine as the anionic binding sites within vacuoles [18,27]. Thus, it is quite possible that most of pclyamines in vacuoles (Table I) are tightly bound with polyphosphates: the m e m b r a n e vesicle fraction still retained about 30 nmol Pi of poiypilosphate per mg protein, although most intravacuolar polyphosphates were released during the preparation of the m e m b r a n e vesicles [20]. Accumulation of polyamines into the vesicles should be due to the combined effect of active polyamine u p t a k e driven by a proton potential and binding of polyamincs with the anionic sites in the vesicles. In o r d e r to discuss the precise energetics of polyamine transport, we will require an analysis by reconstitution of the purified transport system (transporter) into proteoliposomes. T h e r e are very few reports which show the presence of the polyamine transport system in eukaryotic organelles [15,16,28]. Toninello et aL [15,16] reported a s p e r m i n e / p h o s p h a t e co-transport system in rat liver mitochondria. However, the physiological function of this system is obscure. In S. cerevisiae and IV. crassa, vacuoles actually retain polyamines, especially spermidine, u n d e r the conditions examined (Table ! and [18]). In one m u t a n t of N. crassa, whose growth is highly sensitive to putrescine added ~o the medium, the content of putresine in vacuoles was observed to be remarkably increased [29]. Recently we found a culture condition under which polyamine toxicity for growth of the wild-type strain of S. cerevisiae was induced: spermine was the most toxic of the three (Kakinuma, Y., Masuda, N. and Igarashi. K., manuscript in preparation). U n d e r this condition, more polyamines were accumulated by the cells than were required for the growth; the vacuolar poi3,amine level was concomitantly increased. Vacuoles mov serve as a storage system(s) of polyamines to provide a homeostatic reserve or as a detoxification system to prevent the effects of excess free polyamincs in the cytoplasm. The active polyamine transport system reported here possibly functions to facilitate the accumulation of polyamines in vacuoles. Experiments are now in progress to exam-
ine the physiological significance of the vacuolar polyamine transport sysytem in more detail. Acknowledgments We thank Dr. Y. A n r a k u for his encouragement. W e also thank Dr. Y, Ohsumi and Dr, E. Uchida for their valuable discussions. This work was supported in part by a grant-in-aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. References I Cohen, S.S. (1971) Introduction to the polyamines, pp. 1-179, Prentice-HaU, Englewood Cliffs, NL 2 Tabor, C.W. and Tabor, H. (1984) Anna. Rev. Biochcm, 53, 749-790. 3 Pegg, A.E. (1988) Cancer Res, 48, 759-774. 4 Tabor, C.W. and Tabor, H. (1966) J. Biol. Chem. 241, 3714-3723. 5 Pohjanpclto, P. (1976) J. Cell Biol. 68, 512-520. o Kano, -". and Oka, T. (1976) J. Biol. Chem. 251, 2795-2800. 7 DiPasquale, A, White. D. and McGuire, J. (1978) Exp. Cell Res. 116, 317-323. 8 Kashiwagi,K., Kobayashi, H. and lgarashi, K. (1986) J. Bacteriol. 165, 972-977. 9 Kashiwagi, K., Hosokawa, N., Furuchi, T., Kohayashi, H., Sasakawa, C, Yoshikawa, M. and lgarashi, K. (1990) J. Biol. Chem. 265, 20893-20897. 10 Kakiauma, Y., Hoshino, K. and Igarashi, K. (1988) Ear. J. Biochem. 176, 409-414. 11 Kashiwagi,K. and lgarashi, K. (1988) J. BacterioL 170,3131-3135. 12 Igarashi, K., Sakamoto, I., Goto, N., Kashiwagi, K., Honma, R. am1 Hirose, S. (1982) Arch. Biochem. Biophys. 219, 438-443. 13 Kakegawa, T., Sato, E., Hirose, S. and Igarashi, K. (1986) Arch. Biochem. Biophys. 251,413-420. 14 Watanabe, S., Kusama-Eguchi, K., Kobayashi, H. and Igarashi, K. O991) J. Biol. Chem. 266, 20803-20809. 15 Toninello, A., Di Lisa, F., Siliprandi, D. and Siliprandi, N. (1985) Biochim. Biophys. Acla 815, 399-404. 16 Toninello, A., MioUo, G., Siliprandi, D.. Siiinrandi, N. and Garlid, K.D. (1988)J. Biol. Chem. 263,19407-19411. 17 Matile, P. (1978) Annu. Rev. Plant Physiol. 29, 193-213. 18 Paulus, T.L, Cramer. C.L. and Davis, R.H. (1983) J. Biol. Chem. 258, 8608-8612. 19 Ohsumi, Y. and Anra~a, Y. (1981) J. Biol. Chem. 256, 20"/9-2082 20 Kzkinuma, Y., Ohsumi, Y. and Anraku. Y. (1981) J. Biol. Chem. 23f, 10859-10863. 21 Kakinnma, v.. S-~ ~maki,Y., lto, K., Cragoe, E. Jr. and Igarashi, K. (1987) Arch. Biochcm. Biophys. 259, 171-177. 22 Lowry, O.H., Resebrough. N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 23 Ohsumi, Y. and~Anraku, Y. (1983)J. Biol. Chem. 258, 5614-5617. 24 Sato, '1'., Ohsumi, Y. and A~waku, Y. (1984) J. Biol. Chem. 259, 11505-11508. 25 Bowman, F...I. Siehers. A. ahd Altendorf, K. (1988) Proc. Nail. Acad. Sci. USA 85. 7972-7976. 26 Diirr, M., Urech, g., Boiler, T., Wiemken, A., Schweneke,J. and Na~, M. (1979) Arch. MieribioL 121,169.-175, 27 Cramer, C.L. and Davis, R.H. (1984) J. BioL Chem. 259, 51525157. 28 Pistocehi, R., Antognoni, F., Bagni., N. and Zannoni, D. (1990) Plant Physiol. 92, 690-695. 29 Davis. R.H. and Ristow, J.L. (1991) Arch. Biochem. Biophys.285, 306-31 I.
Biochimica ei BiopltysicaActa, 1107(1992) 126-130 © 1992ElsevierScience PublishersB.V.All rightsreserved0005-2736/92/$05.00
Proton potential-dependent polyamine transport by vacuolar membrane vesicles of Saccharomyces cerevisiae Yoshimi Kakinuma, Naoyuki Maqttda and Kazuei Igarashi Facultyof PharmaceuticalSciences, Chiba Unirersity, Chiba (Japan)
(Received21 January 1992)
Key words: Polyaminetransport; Proton potential; Vacuolarmembranevesicle;(S. cereridae) Vacuolar membrane vesicles of Saccharomyces cerevisiae accumulated spermine and spermidme in the presence of ATP, not in the presence of ADP. Spermine and spermidine transport at pH 7.4 showed saturation kinetics with K m values of 0.2 mM and 0.7 raM, respectively. Spermine uptake was competitively inhibited by spermidine and putrescine, but was not affected by seven amino acids, substrates of active transport systems of vacuolar membrane. Spermine transport was inhibited by the H +-ATPasespecific inhibtors bafilomycin A I and N,N'-dicyclohexylcarbodiimide, but not by vanadate. It was also sensitive to Cu 2+ or Zn 2+ ions, inhibitors of vacuolar H +-ATPase. Both 3,5-di.tert.butyl-4-hydroxTbenzilidenemalononitrile (SF6847) and nigericin blocked completely the spermine uptake, but valinomyein did not. [14C]Sp~.rmine accumulatcd in the vesicles was exchangeable with unlabeled spermine and spermidine. However, it was released by a protonoph~re only in the presence of a counterion such as Ca z+. These results indicate that a polyamine-specific transport system depending on a proton potential functions in the vacuolar membrane of this organism.
Introduction Polyamin~.s, represented by putrescine, spermidine and spermine, are found to be widely distributed from bacteria through to higher organisms [1]. The significance of these polyamines in cell proliferation is clear, since the depletion of intraceUular polyamines by the blockage of synthetic pathways with inhibitors or the mutation of polyamine biosynthetic enzymes halts the cell growth. It can then be restored by the addition of polyamines to the medium [2]. The cellular polyamine level is usually regulated by both its synthetic pathway via ornithine decarboxylase as the rate-limiting enzyme with a very short half-life and by its degradative pathway [2]. It has recently become accepted that movements of polyamines across the cell membrane are also important for regulation of its cellular content [3]. Uptake systems specific for polyamines are reported in bacteria and animal cells Correspondence: K. Igarash!, Faculty of Pharmaceutical Sciences, Chiba University,1-33 Yayoi-cho,Chiba 263, Japan. Abbreviations: ANS, l-anilinonaphthalene.8-sulfonicacid; CCCP, carbonyl cyanide m-chlorophenylhydrazone;DCCD. N,N'-dicyclohexylcarbodiimide;HPLC. high-performanceliquid ¢.hromalography; Mes, 2-(N-rnorpholino)ethanesulfonicacid; SF6847, 3,.~-di-tert-butyl4-hydroxybenzilidenemalononitrile.
[4-7]; we found three polyamine uptake systems, which depend on the proton potential, in Escherichia coil [8,9], and a membrane potential-dependent uptake system of polyamines in bovine lymphocytes [10]. As polyamines and their derivatives synthesized in vivo are released into the medium [11], their extrusion pathways must exist. As polyamines easily bind to cellula~ anionic constituents such as RNA, DNA and phospholipids [12], it is hard to define the target site(s) crucial for cellular physiology. However, we understand that binding of polyamine to the ribosomal machinery or m R N A plays an important role in protein synthesis in the cytoplasm [13,14]. The cytoplasmic polyamine level is probably influenced by its compartmentalization in various organelles of eukaryotes: a few reports have indicated the existence of the polyamine transport system in mitochondria [15,16]. Vacuoles compose the largest compartment in plants and fungi [17]. It has been proposed that vacuoles maintain cytoplasmic homeostasis of nutrients and ions and act as a metabolic compartment or lytic compartment in intracellular digestive processes [17]. In Neurospora crassa, it was reported that spermidine is localized in vacuoles [18]. However, the mechanism of polyamine accumulation into vacuoles has ,,ot been investigated.
127 We examined here polyamine transport by vacuolar membrane vesicles of Sa,echaromyces cerevisiae, and found a polyamine specific transport system depending on a proton potential. Materials and Methods
Strain and culture condition. All the experiments were conducted with Saccharomyces cerevisiae haploid strain X2180-1A from the Yeast Genetic Stock Center, Berkeley, CA, USA. Cells were grown in medium (YEPD) containing 1% Difco yeast extract, 2% polypeptone and 2% glucose at 30°C with subsequent shaking at 120 strokes per rain. Chemicals. [t4C]Putrcscin¢, spermidin¢ and spermine (4 G B q / m m o l ) were purchased from DuPontNew England Nuc~.ear. SF6847 was a gift from Dr. Y. Anraku, University of Tokyo, Japan. Other chemicals used were of analytical grade. Preparation of spheroplast lysate, intact vacuoles and right-side-out vacuolar membrane vesicles. Spheroplast lysate, intact vacuoles and right-side-out vacuolar membrane vesicles, which were essentially free of mitochondrial contamination (no succinat¢ dehydrogenas¢ activity as a marker enzyme for mitochondria), were prepared by the method of Ohsumi and Anraku [19]: in short, intact vacuoles were released into the lysat¢ by an 'osmotic shock' disruption of spheroplasts, and then separated from other organelles and cellular constituents by centrifugation. Only intact vacuoles are floated in 8% Ficoll buffer during centrifugation at 2 6 6 0 0 × g for 30 rain (see rgf. 19 for details). The vesicles were perfectly spherical and 0.2-1.0 /~m in diameter. An internal water space of vacuolar membrane vesicles of 5.2 p.I/mg membrane protein [20] was used for calculation of the concentration. Transport assay. The standard assay mixture (100 btl) consisted 6f 25 mM Tris-Mes (pH 7.4), 4 mM MgCI 2, 25 mM KCI, 0.5 mM ATP, 0.1 mM polyamine (74 MBq/mmol) and 30-50 p.g protein of vacuolar membrane vesicles. The reaction at 25°C was initiated by adding labeled substrate and :~rminated by diluting the reaction mixture with 5 mi of cold buffer consisting of 10 mM.Tris Mes (pH 6.9), $ mM MgCI 2 and 25 mM KCI. The vacuolar membrane vesicles were recovered on a membrane filter (Millipore, 0.45/tm) and washed with 5 ml of the above buffer. The radioactivity was determined with a liquid scintillation counter. Polyamine analysis. Polyamine contents in spheroplast lysates, intact vacuoles and vacuolar membrane vesicles were determined with HPLC as described elsewhere [21] after extraction with hot triehloroacetic acid, Other procedures. Protein was determined by the method of Lowry et al. [22] with bovine serum albumin as standard.
TABLE 1 Retentionof polyamim's it! cacuoles S. cerevisiaeX2180.1Agrown on YEPD mediumwere harvestedat 3' 107cells/ml, and then spheroplastlysateand intact vacuoleswere prepared as described in Materials and Methods. Intact vacuoles were then disrupted by osmotic shock to convert to membrane vesicles, and the vesiclesand their supernatant were recovered by centrifugation[191.These fractionswere extracted twice with 0.5 ml of 5% Iric~ ...'oacetic acid, and aliquots were then analyzed for polyaminesby HPLC as described previously[21]. Standard error was within the range of 5% in triplicate experiments. PUT, putrescine; SPD, spermidine; SP. sperrnine. Numbers in parentheses show total polyaminecontent (in nmoDin each f:'sction: n.d., not detected. Fraction
Pmtein nmol/mgpmtein(nmol) (mg) PUT SPD SP
Sphcroplast lysate
238
Intact vacuoles 1.7 After precipitation Precipitate(vesicles) 0.4 Supernatant 1.5
3.9
51.6
0.8(1.4) ~.6(103) n.d, 1.1(I.7)
0.5 1.9(3.2)
72.5(~) 0,9(0.4) 43.3(65) 1.9(2.9)
Results
Retention of polyamines in vacuoles Table I shows the polyaminc contents of spheroplast lysate, intact vacuoles and vacuolar membrane vesicles. Strain X2180-1A was grown on YEPD medium, harvested at 3" 107 cells/ml, and individual fractions were prepared as described in Materials and Methods. In the fraction of purified vacuoles, putrcscine, spermidine and spermine were all detected, although spermidine was preferentially presented. By osmotic disrup, lion, 100% of putrcsin¢, 65% of sperrnidin¢ and 90% of spermine in the vacuoles were released, indicating that all these polyamines were actually retained in vacuoles. Based on a-mannosidase activity as a specific marker enzyme of the vacuolar membrane [20], about 13% of total vacuoles in the spheroplast lysate are recovered by this purification method. This means that 1% of putrescine, 6% of spermidine and 21% of spcrmine of the lysat¢ are localized in vacuoles although the possibility of a leakage of polyamine from vacuoles by this osmotic shock method cannot be ruled out (Table 1): it was reported that about 30% of cellular spermidine is localized in vacuoles of Neurospora crassa
[181. ATP-dependent polyamine uptake by vacuolar mere. brahe ve,~icles Anraku and his colleagues have been extensively studying the energetics of yeast vacuoles, and have demonstrated ATP-driven transport systems of Ca 2+ and amino acids [19,23,24]. All of these systems are actually driven by a proton potential (about 180 mV,
128 TABLE II 10!!
.
.
.
.
.
.
.
.
.
Kineticparametersof A TP-rh,pendent pnlyammeuptake
.
Uptake assayswere performed as described in Materialsand Methods. Kinetic constants were estimated from the initial velocityof uptakes.The Ki valuesof competitiveinhibilkmweredeterminedby varying the substrate concentrafion in the presence of a fixed inhibitoreonceutration.Standard error was will|inthe rangeof I(1% ill
iF
triplicate experiments. SP, sperminc; SPD, spermidine; PUT, pu.
trescine. Substrate
Km (mM)
Putrescine Spermidinc Spermine
0
0
"
"
2
"
' 4
6
8
10
(rim)
Fig. I. Time courseof ATP-dependent polyamineuptakeby wteuolar membrane vesicles.Uptakes of spermine (el. spermidine ( • ) and putreseine (Ill) into the vesicleswere assayed under standard conditions in the presence of 11.5 mM ATP. Each point represents the average of triplicate experiments. Standard error was within the range of I()q~:.
intravaeuolar positive) generated by a vacuolar H +translocating ATPase, and are suggested to be an n H + / C a -'÷ antiporter or n H + / a m i n o acid antiporters [19,23,24]. We examined the effect of ATP on accumulations of [t4C]putrescine, spermidine and spermine into vacuolar membrane vesicles (Fig. I). The uptake activities of spermidine and spermine were greatly stimulated by ATP, but the activity of ATP-dependent putresine uptake was very low. In the absence of ATP, no significant uptake of polyamines was observed. At 5 min after the addition of ATP, established concentration gradients of spermidine and spermine across the vacuolar membrane were approximately 4 and 17, respectively, indicating ATP-dependent active transport of these polyamines by vacuolar membrane vesicles. When the experiments were performed at 0°C, no accumulation of these polyamines was observed even in the presence of ATP.
Specificio of the uptake system Table 11 shows the kinetic parameters of ATP-dependent polyamine uptake by vesicles. Considering the values of K m and Vma~ for the uptake activities of putreseine, spermidine and spermine (Fig. 1), spermine was the substrate with the highest affinity. Spermidine uptake was inhibited by putrescine (data not shown) and spermine, and spermine uptake was inhibited by spermidine and putrescine. Inhibitions by these
2.0 0.7 0.2
Vma~ (nmolmin t mg protein)- i 0.8 6 7
Ki (mM)
0.1 (SP) 0 8 (SPD), 1.1 (PUT)
polyamines were apparently all competitive; the K~ values of these polyamines are shown in Table ll. Eight transport systems of amino acid are found in vacuolar membrane [24]. Seven of these are eriven by a proton potential generated by H+-ATPase [20]. To determine whether polyamine uptake is mediated by a specific transport system distinct from these systems, the effect of these seven amino acids on spermine uptake by ve~eles was examined (Table liD. Sperm(he uptake was hardly inhibited by the addition of these amino acids, suggesting that an ATP-dependent transport system specific for polyamines exists on the vacuolar membrane.
Properties of polyamhw uptake Table IV shows the nucleotide specificity of spermine uptake. Spermine transport was not only stimulated by ATP but also by GTP, UTP and CTP in this order. No transport of spermine was observed by the addition of ADP. This preference corresponded well to the nueleotide specificity of the vacuolar membrane H+-ATPase [20]. The effect of various ATPase in-
TABLE III
I'~ffi'clof rut(pus L-ambzoachlson ATP.dept'ndentspernlbwuptake The initial rates of sperm(he uptake were determined under star,. dard condithms. The concentrationsof labeled spermine and unlabeled amino acid were 0.1 and 2 raM. respectively.Each point represents tbs mean of duplicate experiments. Aminoacid
Relative activily (%)
None Arginine Lysinc Histidinc Plmnylalanine Tyrosiue Glutamine lsoleucine
I(}0 102 83 138 99 84 l Ir~ 127
129 TABLE IV NucleotMe.~pecificityof spermine uptake Initial rates of spermine uptake were determined uuder standard conditions.The concentrationof nucleotide was 0.4 raM. Each point represents tile mean of duplicate experiments.
GTP UTP CTP ADP
I:'
.
.
.
.
.
•
.
.
.
.
°
o
I.
86 45 32 0
i,
hibitors and ionophores on ATP-dependent spermine uptake was also examined (Table V). Vesicles preincubated with 10 p.M bafilomycin A t [25] or 0.1 mM N,N'-dicyclohexylcarbodiimide, a specific inhibitor for the vacuolar H +-ATPase, showed no spermine uptake activity. By contrast, vanadate, an inhibitor of the ErE 2 type ATPase, slightly activated spermine uptake. Cupric and zinc ions, which strongly inhibit H+-ATPase [20], blocked the activity completely. Calcium ions also inhibited the uptake, but to a lesser degree than Cu 2+ and Zn 2+. When magnesium ions were removed from the reaction mixture, the uptake activity was completely, lost, indicating that they are essential for the uptake activity. These results suggest that the vacuolar [-I+-ATPase participates in the spermine uptake by membrane vesicles. Moreover, spermine uptake was severely inhibited b~ the protonophores SF6847 and CCCP, as well as by an electroneutral K + / H + an-
TABLE V
Effi'ct of carious reagentson ,4 TP.dependent Sl~'nnhwuptake Assayswere p~:rformedunder standard conditions and the activity without addition (control) w::s taken as 100%. Membrane vesicles were preincuhatedwith bafilomyeinA I or DCCDat 25°Cfor 10 rain before assay. Other rcageats were added just before assay. Each point representsthe mean of duplicate experiments.
Addition
1 0 • A.
Cunceati'ation (mM)
Relativeactivity (%)
0.01 O.2 0.1
9 0 152
MgCI~
0.I
I0{}
CaCI: CuCI2 ZnCI2
0.1 0.1 0.1
55 0
~
(ml,)
Fig. 2. Effluxof [t4Clspcrminefromvacuolarmembranevesicles.(A) Effect of ur.labeled polyamine.ATP-drivenspermineuptake (o) was carried out un..ter standard conditionswith 0.1 mM 114C]spermine. At the arrow, 1 mM spermine to) or 2 mM spermidine (Ill) was added to portionsof the reaction mixture.(B) Effect of 81:6847 and Ca:+, ATP.driven spermineuptake (0) was carried out as the same as in Fig. 2A. At the arrow.2/zM SF6847(o) or SF6847plus 2 mM CaCI2( • ) was added to portionsof the reaction mixture.
tiporter, nigericin, indicating that it depends on a proton potential. However, valinomycin did not inhibit spermine uptake activity (Table V). The same effect of these reagents on ATP-dependcnt spermidine uptake was obtained (data not shown). All these results suggest that active polyamine transport by membrane vesicles is driven by a proton potential generated by the vacuolar H %ATPase. To determine whether accumulated spermine in vesicles is in a free, osmotically active form, the effiux of spormine was examined (Fig. 2). The spermine accumulated in the vesicles was exchangeable with a 10-fold excess of nonradioactive spermine (Fig, 2A). Spermidine also prompted the exchange diffusion of spermine, but to a lesser extent. Howew:r, spermine was not in a completely free form, since SF6847 did not readily release it (Fig. 2B). When Ca 2+ ions were added as eonv.ter ion to the suspension, the effiux of spermine became accelerated (Fig, 2B).
ATPase inhibito~
BafilomycinA t DCCD Sodiumvanadate Divalent cations
Ionophores SF6847 CCCP
0
0.001
12
Nigericin
O,D2 0.O01
6 23
Valinomycin
0.01
98
Discussion
In this work, we used vacuolar membrane vesicles in which the intravacuolar polyamines were depleted (Table I), instead of the fragile vacuoles, and found a new active transport system of polyamines in vacuoles of S. cerevisiae. This system was ATP-dependent and polyamine-specific. Judging from the sensitivity of spermine uptake to various ATPase inhibitors and ionophores, this system depends on a proton potential generated by H+-ATPase [20]. We expect that the polyamine uptake system may be a substrate/nH +
130 antiporter as reported in the uptake systems of arginine and Ca 2+ [19,23]. However, we must consider that accumulated spermine in vesicles is probably not in a free form; spermine at a steady-state level was not released even when the proton potential was dissipated by the addition of the protonophore SF6847 (Fig. 2B). Accumulated sperreine probably binds with some anionic constituents in the vesicles. Diirr et al. [26] measured the amounts of amino aCids and polyphosphates in the vacuoles of yeast, and found a linear correlation of their contents, from which they suggested that polypbosphates act as ion exchange resins to conserve amino acids stably in the vacuolar space [26]. Also in the case of N. crassa, it is pointed out that polyphosphates sequester spermidine as the anionic binding sites within vacuoles [18,27]. Thus, it is quite possible that most of pclyamines in vacuoles (Table I) are tightly bound with polyphosphates: the m e m b r a n e vesicle fraction still retained about 30 nmol Pi of poiypilosphate per mg protein, although most intravacuolar polyphosphates were released during the preparation of the m e m b r a n e vesicles [20]. Accumulation of polyamines into the vesicles should be due to the combined effect of active polyamine u p t a k e driven by a proton potential and binding of polyamincs with the anionic sites in the vesicles. In o r d e r to discuss the precise energetics of polyamine transport, we will require an analysis by reconstitution of the purified transport system (transporter) into proteoliposomes. T h e r e are very few reports which show the presence of the polyamine transport system in eukaryotic organelles [15,16,28]. Toninello et aL [15,16] reported a s p e r m i n e / p h o s p h a t e co-transport system in rat liver mitochondria. However, the physiological function of this system is obscure. In S. cerevisiae and IV. crassa, vacuoles actually retain polyamines, especially spermidine, u n d e r the conditions examined (Table ! and [18]). In one m u t a n t of N. crassa, whose growth is highly sensitive to putrescine added ~o the medium, the content of putresine in vacuoles was observed to be remarkably increased [29]. Recently we found a culture condition under which polyamine toxicity for growth of the wild-type strain of S. cerevisiae was induced: spermine was the most toxic of the three (Kakinuma, Y., Masuda, N. and Igarashi. K., manuscript in preparation). U n d e r this condition, more polyamines were accumulated by the cells than were required for the growth; the vacuolar poi3,amine level was concomitantly increased. Vacuoles mov serve as a storage system(s) of polyamines to provide a homeostatic reserve or as a detoxification system to prevent the effects of excess free polyamincs in the cytoplasm. The active polyamine transport system reported here possibly functions to facilitate the accumulation of polyamines in vacuoles. Experiments are now in progress to exam-
ine the physiological significance of the vacuolar polyamine transport sysytem in more detail. Acknowledgments We thank Dr. Y. A n r a k u for his encouragement. W e also thank Dr. Y, Ohsumi and Dr, E. Uchida for their valuable discussions. This work was supported in part by a grant-in-aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. References I Cohen, S.S. (1971) Introduction to the polyamines, pp. 1-179, Prentice-HaU, Englewood Cliffs, NL 2 Tabor, C.W. and Tabor, H. (1984) Anna. Rev. Biochcm, 53, 749-790. 3 Pegg, A.E. (1988) Cancer Res, 48, 759-774. 4 Tabor, C.W. and Tabor, H. (1966) J. Biol. Chem. 241, 3714-3723. 5 Pohjanpclto, P. (1976) J. Cell Biol. 68, 512-520. o Kano, -". and Oka, T. (1976) J. Biol. Chem. 251, 2795-2800. 7 DiPasquale, A, White. D. and McGuire, J. (1978) Exp. Cell Res. 116, 317-323. 8 Kashiwagi,K., Kobayashi, H. and lgarashi, K. (1986) J. Bacteriol. 165, 972-977. 9 Kashiwagi, K., Hosokawa, N., Furuchi, T., Kohayashi, H., Sasakawa, C, Yoshikawa, M. and lgarashi, K. (1990) J. Biol. Chem. 265, 20893-20897. 10 Kakiauma, Y., Hoshino, K. and Igarashi, K. (1988) Ear. J. Biochem. 176, 409-414. 11 Kashiwagi,K. and lgarashi, K. (1988) J. BacterioL 170,3131-3135. 12 Igarashi, K., Sakamoto, I., Goto, N., Kashiwagi, K., Honma, R. am1 Hirose, S. (1982) Arch. Biochem. Biophys. 219, 438-443. 13 Kakegawa, T., Sato, E., Hirose, S. and Igarashi, K. (1986) Arch. Biochem. Biophys. 251,413-420. 14 Watanabe, S., Kusama-Eguchi, K., Kobayashi, H. and Igarashi, K. O991) J. Biol. Chem. 266, 20803-20809. 15 Toninello, A., Di Lisa, F., Siliprandi, D. and Siliprandi, N. (1985) Biochim. Biophys. Acla 815, 399-404. 16 Toninello, A., MioUo, G., Siliprandi, D.. Siiinrandi, N. and Garlid, K.D. (1988)J. Biol. Chem. 263,19407-19411. 17 Matile, P. (1978) Annu. Rev. Plant Physiol. 29, 193-213. 18 Paulus, T.L, Cramer. C.L. and Davis, R.H. (1983) J. Biol. Chem. 258, 8608-8612. 19 Ohsumi, Y. and Anra~a, Y. (1981) J. Biol. Chem. 256, 20"/9-2082 20 Kzkinuma, Y., Ohsumi, Y. and Anraku. Y. (1981) J. Biol. Chem. 23f, 10859-10863. 21 Kakinnma, v.. S-~ ~maki,Y., lto, K., Cragoe, E. Jr. and Igarashi, K. (1987) Arch. Biochcm. Biophys. 259, 171-177. 22 Lowry, O.H., Resebrough. N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 23 Ohsumi, Y. and~Anraku, Y. (1983)J. Biol. Chem. 258, 5614-5617. 24 Sato, '1'., Ohsumi, Y. and A~waku, Y. (1984) J. Biol. Chem. 259, 11505-11508. 25 Bowman, F...I. Siehers. A. ahd Altendorf, K. (1988) Proc. Nail. Acad. Sci. USA 85. 7972-7976. 26 Diirr, M., Urech, g., Boiler, T., Wiemken, A., Schweneke,J. and Na~, M. (1979) Arch. MieribioL 121,169.-175, 27 Cramer, C.L. and Davis, R.H. (1984) J. BioL Chem. 259, 51525157. 28 Pistocehi, R., Antognoni, F., Bagni., N. and Zannoni, D. (1990) Plant Physiol. 92, 690-695. 29 Davis. R.H. and Ristow, J.L. (1991) Arch. Biochem. Biophys.285, 306-31 I.
